Você está na página 1de 41

The Evolution of Elliott Carter's Rhythmic Practice

Author(s): Jonathan W. Bernard


Source: Perspectives of New Music, Vol. 26, No. 2 (Summer, 1988), pp. 164-203
Published by: Perspectives of New Music
Stable URL: http://www.jstor.org/stable/833189
Accessed: 07/02/2010 18:10

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=pnm.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

Perspectives of New Music is collaborating with JSTOR to digitize, preserve and extend access to Perspectives
of New Music.

http://www.jstor.org
THEEVOLUTION
OF
ELLFTOTT
CARTER'S
RHYTHMIC
PRACTICE

W. BERNARD
JONATHAN

INTRODUCTION

ELLIOTT CARTER'SWORKover the past forty years has made him per-
haps the most eminent living American composer, and certainlyone of
the most important composers of art music in the Westernworld. This span
of time does not encompass his entire careerto date-hardly surprisingfor a
man who will soon celebratehis eightieth birthday-but in crucialrespects
deserves to be treated separatelyfrom the period which preceded it. Car-
ter's earliestpublished works, dating from the late 1930s, are not juvenilia,
but they are cast in a ratherconventional mold; his music from this period
has that neoclassical, "Americanized" sound which is readily recognizable
as a feature of the style adopted then by many a former Boulanger student.
In the late 1940s, however, Carterstruck off in a startlinglynew direction.
ElliottCarter'sRhythmicPractice 165

In retrospect, the Piano Sonata of 1945-46 was a harbingerof change, but


the first work which actuallysoundsdistinctly differentis the Cello Sonata of
1948, Carter'sfortieth year.
What had happened? It is true that much of Carter's musical education
had been quite traditional,even conservative. But the same can be said, for
instance, of Charles Ives, who acquired his formal training from Horatio
Parker. At least as important to Ives's eventual development were much
earlier influences, notably his father the bandmaster and musical free-
thinker. The sources of Carter's nonacademic education lay not quite so
close to home, but in his case they didn't need to, for Carter's circum-
stances during his formative years were distinctly enviable: his family lived
in New Yorkand made frequent trips to Paris. From an earlyage he was fas-
cinated by modern music and heard, by his own account, as much as he
could manage.
Of the many musicalexperienceswhich can now be assessedas influences
upon Carter, one in particular,in light of his course since 1948, stands out.
This is his exposure to the work of three composers who are now well
known for their innovative work in rhythm: Ives, Henry Cowell, and Con-
Ion Nancarrow. Each affected Carter in different ways; all three are men-
tioned in his 1955 article, "The Rhythmic Basis of American Music."'
Here Carterenumerateswhat he saw as Ives's contributions: (1) superposi-
tion of different speeds notated in a common unit; (2) notated rubati
played againststrict time; (3) unrelated levels heard simultaneously, includ-
ing the use of a softly played backgroundthat could be heard distinctly only
during the silences between fragments of louder music, as in The
Unanswered Question.2In the case of Cowell the influence came by way of
his book, New Musical Resources,published in 1930. Carter credited the
rhythmic techniques described therein with having "furnished me with
many ideas."3 As for Nancarrow, Carter knew at least three of the pieces
later collected under the title StudyNo. 3, as well as StudyNo. 1; these, pre-
sumably, he saw while serving on the editorial board of New Music Quar-
terlyduring the 1940s (StudyNo. 1 was published there in 1951).4 In "The
Rhythmic Basisof American Music" Cartercites a passagefrom StudyNo. 1
which combines "four distinct planes of rhythm."5
But exactly what form did this influence take?Certainly not a desire to
adopt others' methods wholesale; for Carter, all three approachesto rhyth-
mic emancipation proved, in one way or another, unsatisfactory. Ives's
innovations, for instance, for all the initial excitement they provoked, even-
tually appeared to him deeply problematic. Beginning in the late 1930s,
Carter'sattitude toward Ives's music was ambivalent. For example, he has
described the orchestral music as consisting of "very large amounts of
undifferentiatedconfusion... during which many conflicting things hap-
pen at once without apparent concern either for the total effect or for the
166 of NewMusic
Perspectives

distinguishabilityof various levels."6 In a review of the New Yorkpremiere


of the ConcordSonata in 1939 for the periodicalModernMusic, Cartercasti-
gated the work for what he felt were its various shortcomings, including its
muddled rhythmic character.7Cowell's ideas were difficult, if not impossi-
ble, to implement using conventional notation; Carter found himself
unwilling to contemplate the radical revisions in notation that Cowell
advocated, such as new note shapes to representnew divisions of the meas-
ure.8 He became suspicious besides of the focus of Cowell and others on
"purely physicalpossibilitiesand their juggling" in their efforts to organize
time.9
Here Carter's interest in exploration of new rhythmic territory appears
tempered by the concerns of a composer oriented toward more or less tradi-
tional performance. From his dissatisfactions with Ives's methods Carter
learned that the introduction of new rhythmic intricacies-specifically, the
simultaneous projection of "unrelated" patterns, or patterns related only
by complex ratios-posed enormous problems of control, but he did not
want to give up writing for human musicians, as Nancarrow, with his
player pianos, had done from the 1940s on; nor did he care for the idea of
using a machine like Cowell's rhythmicon, even in combination with an
orchestra, as Cowell had done in his work Rhythmicana(1931). There can
be little doubt, however, that Carter's exposure to these attempts to
develop radicallynew rhythmic methods impressed him, to the point of
convincing him that he could find a solution of his own. The record of his
compositions from 1948 on shows no rapid or radical shift in technique;
instead, Carter moved gradually, absorbing the lessons afforded by his
experience with each new work. His progress along the path to the works
of the early 1960s is quite clear; nearly every successive composition from
the Cello Sonata (1948) to the Double Concerto (1961) represents a sub-
stantial change in the treatment of rhythmic matters.

DEFINITIONS, GENERALITIES, AND AN OVERVIEW

In this essay, the term "simultaneity" refersto the projection during some
passage in a composition of two or more distinctly different rhythmic pat-
terns, often taking the form of different speeds(see below) and usually
occurring as separate strands or parts of the musical texture. By "succes-
sion" is meant the temporal arrangementof two or more different rhyth-
mic patterns such that one pattern is followed by a second (and the second
by a third, and so on) as the music progresses, either within the same part
or from one part to another, with the first pattern either being abruptly
superseded by the second or-as in the case of metric modulation, to be dis-
cussed-undergoing a transformation or a series of transformations that
graduallyproduces the second rhythmic pattern from the first.
ElliottCarter'sRhythmicPractice 167

Well before 1948, Carter's acquaintancewith the work of Ives, Cowell,


and Nancarrowbrought him to a realizationthat, in its recounting, sounds
almost epiphanic:

Around 1944... I suddenly realized that, at least in my own educa-


tion, people had alwaysbeen consciously concerned only with this or
that peculiar local rhythmic combination or sound-texture or novel
harmony and had forgotten that the really interesting thing about
music is the time of it-the way it all goes along.'0

More specifically,he has stated: "I began as early as 1944, in works like my
Holiday Overture,to think in terms of simultaneous streams of different
things going on together."" Judging from the significance which Carter
himself, in "The Rhythmic Basis of American Music," attributed to the
work of his predecessors, it does follow naturallythat Carter should have
been impelled to attempt the realization of simultaneityin his own music.
But in charting the evolution of his new rhythmic practice, especially at its
beginning, one must distinguish between the general musical problems
that engaged Carter's attention and the specific compositional solutions
devised for them. The degree to which the HolidayOverture,for example,
reflects this fascination with "simultaneous streams of different things
going on together" is limited by the techniques then at his disposal. The
simultaneity, where it exists at all, is of only a ratherrudimentarysort. (See
Example 1.)
In his recent book on Carter's music, David Schiff asserts that "in the
late 1940s Carter developed a new rhythmic language" but then proceeds
to discuss it as though all its essential features were alreadyin place by the
time of the Cello Sonata.12In taking this approach, Schiff seems to dis-
regardan essential point: that Carterat the beginning of this evolutionary
period was more preoccupied with techniques of rhythmic successionthan
with those of simultaneity. Schiff is right, though, to remind his readersof
the importance of simultaneity to an understanding of what Carter was
working toward from the late forties on, for it was precisely through refin-
ing the mechanisms of rhythmic succession that Carter eventually solved
the problems posed by simultaneity. Simultaneity emerges in an
increasinglyintegralrelationshipto the substanceof the works after1948, as
Carter'sconfidence in his handling of rhythmic-successivetechniques grew.

CELLOSONATA(1948)

Carter'sfirst great stride in the direction of controlling large-scalerhythmic


structure was metric modulation, as it has now come to be called.13This
168 of NewMusic
Perspectives

#_MI- IL
(W ^
\n^JJrj^^^ ; nrf^
$ z jS Pool
Winds ] IVf leggiero
M2 . _1 .,I
_
t-
.
. ho.u ----A-- - t
-

Si

s1~~i ; # -d I -
#,;
:-0, -
#! Lp-
Strings
-0.~~ 0&

V I
p espress.
cantandosost.

(etc.)

jfJ4S#J; #S #: g

- rK (etc.)

EXAMPLE 1: Holiday Owrture, MM. 103-107

technique was introduced in the Cello Sonata and has been a fixture in his
work ever since; it provides a way of moving from one speed to another by
means of changes of time signature and redivision of the beat.14The speed
changes often take place in several steps and are carefully prepared,
although they may occur fairly gradually or fairly quickly. In Example 2
(Cello Sonata, III, measures 6-13), the initial value of = 70 is trans-
formed into h = 60. Below the music is indicated the series of steps which
produces this transformation.
Metric modulation, then, although it does produce momentarily over-
lapping speeds, is really a device for composing a seriesof different speeds,
each precisely related to its immediate predecessor and its immediate suc-
cessor. Carterhas said that he did not venture into polyrhythms before the
Cello Sonata because of a "desire to remain within the realm of the per-
formable and auditorily distinguishable divisions of time. ..."'5 He does
not venture very far in the Cello Sonata either; but even this relativelycau-
tious beginning forecasts future developments, for the Cello Sonata is the
ElliottCarter'sRhythmicPractice 169

=70 = 70 ==i ? ! =60


-6 . .
o , ~ ~ 1 .l .. l

EXAMPLE 2: CELLO SONATA, III, MM. 6-13


170 of NewMusic
Perspectives

first work in which the idea of "simultaneous streams" is realized to any


appreciableextent. Carterhimself was somewhat surprisedby the form that
this work took; twenty yearslater he told an interviewer, while speaking of
the maintenance of distinct charactersfor the piano and cello in the Cello
Sonata and of the generative importance of this distinction to the whole
work, "I don't know how this conception ever originated; it was actually
the first time I ever had this idea."'16What is known is that, typically for
him, the immediate impetus came from thinking about a specific composi-
tional problem-the relationship between two very different instruments-
and the particularnature of the drama of performance that would result
from combining them.
Nevertheless, in the Cello Sonata the feature of simultaneous contrast is
not exactly pervasive;it is evident reallyonly at the beginning and the end
of the work. The fact that the four movements were composed in the order
2, 3, 4, 1 suggests that the manifestations of simultaneity peculiar to the
Cello Sonata found their way into the piece as Carterwas working on it.17
Carterseems to have sought a flexible system, one that would allow him to
express similarity as well as difference and that would allow him to move
with relativeease back and forth between the two. Beginning at measure20
in the first movement, the strict contrast established at the outset between
the two instruments begins to show some signs of erosion: offbeat attacks
in the right hand are set against the "clocklike ticking," as Carterhas called
it, with which the piano began the movement and which now continues in
the left hand. In measure 27 the two hands unite in syncopation, both
against the beat and against the cello's own syncopated gestures. Example
3(a) shows what happens next: the first metric modulation of the first
movement produces the first systematic disruption, in which the piano
presents the polyrhythm 5:4 (literally established at measure 35). This
leads, after a temporaryresumption of the mechanicalquartersof the open-
ing (measures41-48), to a syncopated passage in which the piano actually
imitatesthe characteristicsof the cello part (Example3(b): measures49-54,
and until measure 57).
It should also be pointed out that in this passage two features of simul-
taneous contrastimportant in Carter'slaterworks are not to be found. One
is simultaneous speeds that are reallyspeeds-that is, truly regularized-and
actually perceivableas such. The cello at the opening presents long notes,
espressivo,but at no consistent speed, just in contrast with the perfectly reg-
ular quarters at MM 112 in the piano. The other missing feature is a per-
vasive, structuralconnection of rhythmic contrast to the domain of pitch.
In the later works, specific speeds become associated with specific pitch
entities, but in the opening of the Cello Sonata the starkrhythmic contrast
is mitigated by the pitch structure. Example 4 shows that the first part of
the opening cello phrase can be interpreted as overlapping instances of
ElliottCarter'sRhythmicPractice 171

(a) mm. 33-41

t -- I- .- . . I
(b) mm. 49-54 n J) I - -, K I

semprcJ
(nTf
't
lL 6W;e t1qL tj- ^

EXAMPLE 3: CELLO SONATA, I


172 of NewMusic
Perspectives

EXAMPLE4: CELLO SONATA, I, MM. 1-15

pitch-classsets 5-11 [02347], 4-14 [0237], and 4-17 [0347], with 4-14 and
4-17 each expressed once as literal subsets of 5-11. These sets, as pointed
out in the example, have several prominent correspondents in the piano
part, including-again-literal subset relations. There are also two instances
of 4-17 that link the piano and cello parts. (This analysis hardly accounts
for everything in the pitch domain, but clearly the correspondence indi-
cated is an important part of the complete picture.)18
The third movement of the Cello Sonata provides an excellent illustra-
tion of Carter's early use of metric modulation. As Carter has said, metric
modulation can be used "both as a means of proceeding smoothly or
ElliottCarter'sRhythmicPractice 173

abruptly from one speed to another and as a formal device to isolate one
section from another. "19 This statement carriesthe clear implication that
metric modulation may operate over larger spans of time than the local
connection, which is exactlywhat happens here. The aggregateeffect of the
changes in tempo-all of which are controlled by metric modulation-is to
produce a permanent supersession of the opening speed by another speed
whose first expression in the movement occurs in close relationship to the
opening. This process can be traced through the occurrences of rapid
figuration, which from the beginning of the movement have an obvious
thematic significance(see Example 5). From the sixty-fourth notes in meas-
ure 2, which at a notated tempo of = MM 70 move at the speed of MM
560, the figuration shifts to sextuplet thirty-seconds (measure 5). Example
2 has already illustrated the stages by which these sextuplets (MM 420)
become septuplets at ~ = MM 60 (also MM 420). In every ensuing metric
modulation but one it is the rapid figurationwhich serves as the link, con-
tinuing at the same actual speed across the notated metrical and tempo
changes. Example 6 tabulates these changes and matches them with the
speeds of the rapid figuration at each point.
The significance of this information is, simply, that every one of the
speeds shown can be expressed as a multiple either of 70 or 60 or both.
Thus, while the notated tempi 80, 40, and 48 do actuallyexist, nonetheless
together with 70 and 60 they portray a gradual move from speeds based
upon the former to speeds based upon the latter. The very first figurational
speed, 560, is a multiple of 70 only; the next, 420, is a multiple of both 70
and 60; 560 and 280 are 70-based; and 240 and 480, which end the move-
ment, are multiples of 60 but not of 70. As if to confirm the decisive nature
of this metamorphosis, the final speed, 480, also serves as a specific link to
the fourth movement, which sets forth sixteenths at J = 120.20 The third
movement of this work is, if not the earliest, at least one of the earliest

EXAMPLE 5: CELLO SONATA, III, MM. 1-3


174 of New Music
Perspectives

speed of
measure(s) notated tempo figuration

1 ~ =70 560
5-6 420
10-11 -=60 420

37,40,44-45 =80 560


48 =70 560
49 = 70 560
62 280
63 280
64 280
65 240
73 J =48 240
79-80 480

EXAMPLE 6: CELLO SONATA, III, FIGURATIONALSPEEDS

instances in Carter'smusic of relationshipscreated by multiples of "basic"


speeds. MM 280 and 560 can thus be taken to nprsent 70; 240 and 480 to
represent 60; and 420, depending upon context, to represent either or
both. Representationby multiple continues from this point forward to be
of considerable significance in Carter's schemes of rhythmic relationships,
and it assumes an absolutely crucialrole in the works of the 1960s, where
simultaneity finally comes into its own.
In the Cello Sonata, however, the speed relationshipsare still mainly in
the service of succession. The relationship of the cello to the piano is actu-
ally rathertraditionalin the third movement, and although this means that
their charactersare to some extent contrasted, the figurational common
ground is frequentlyenough expressed to leave no doubt that, rhythmically
speaking, the two parts move together and behave more or less alike.

FIRSTSTRING QUARTET(1951)

In light of Carter's previous works, perhaps the most


striking features of
ElliottCarter'sRhythmicPractice 175

the First String Quartet are its extensive employment of the idea of simul-
taneity and its strong expression of a tendency toward simultaneous differ-
ent speeds. Schiffhas pointed out examplesof rhythmic stratificationin this
work, a practicewhich often involves the simultaneous presentationof two
or more "themes." The first clear instance of such stratification for the
entire ensemble begins at measure 22 in the first movement (see Example
7). As the annotations show, even though the notated tempo is the same in
all parts, the actualspeedsare all different.

(MM120)

MM 48
EXAMPLE 7: FIRST QUARTET, I, MM. 22-29
176 of NewMusic
Perspectives

This kind of procedure is new to Carter'swork, but its applicationis still


fairlylimited at this point in his development. The rhythmic situation is in
rapid flux: alreadyin measure 27 the set of simultaneous speeds, not even
present in its entirety until the entrance of the viola in measure 25, has
alreadybegun to dissolve as the cello shifts from MM 120 to MM 48, or
half the speed of the second violin. By measure 31 the entire scheme has
mutated, and by measure 35 any consistent sensation of simultaneity has
been effectively dispersed. This treatment is typical of the use of simul-
taneous speeds in the First Quartet: their establishment is usually a matter
of only a few measures, and quite often not all parts are included. In fact,
there is only one extended passagein the first movement composed accord-
ing to simultaneously deployed speeds: measures 312-50, which occur at
the very end of the Fantasiaand segue into the Allegro scorrevole.21Even
this passage, however, has its successive aspects, for over the previous
twenty or so measures the speeds to be combined accumulate. Refer to
Example 8, which provides a rhythmic reduction of this passage. By meas-
ure 295 the speed MM 48 has alreadybegun to be established in the viola,
at first as an alternation between MM 24 and 48, as shown (measures
295-300), then uninterruptedlyas MM 24 (measures300-305), then basi-
cally as MM 48 from measure 306 on, with occasionalinterjectionsof MM
24. Meanwhile, MM 180 occurs as two-beat triplets in the other three parts
(measures 296-300), then MM 300 as two-beat quintuplets, again in all
three parts. Triplets (MM 180) resume at measure 304 and become quarters
at the same speed from measure 306 on. Thus, MM 135 in violin II is the
only speed of the four that is actually introduced in the passage of simul-
taneity. At measure 338 the scheme begins to disintegrate as the viola
undergoes a series of accelerative steps: MM 60 (measure 338); MM 80
(measure341); MM 90 (measure342); MM 120 (measure344); and finally
MM 180 (measure350). Over the four measuresbeginning at measure 347
the other three instruments, one by one, defect.
In the case of the passageat measures22-29, alreadyquoted as Example
7, the role of succession in generating the scheme of speeds is also quite
clear, although the method differs somewhat from that used at the end of
the Fantasiamovement. Here it is the rhythmic characteristicsof the open-
ing cello cadenza-specifically, the quintuplet and the seriesof dotted quar-
ters that immediately succeed the first appearanceof the quintuplet-that
serve as the source of multiples and fractions which become the set of
simultaneous speeds (Example9). The first of these speeds to be generated
is MM 96, in violin II at measures 12-15, with one hiatus at measure 14.
This is developed from the cello dotted-quarters (MM 48) by the simple
relationship of 2:1. Next, the quintuplet, only an isolated grouping in
measure 9, becomes a steady stream of notes at MM 360 in measures14-18
ElliottCarter'sRhythmicPractice 177

and serves as the common denominator for the first series of metric modu-
lations in the piece. Its subsequent reductions in speed by one-third, then
by one-half (still in the cello) provide a clue to the significanceof the speeds
of MM 36 (violin I, measure 22) and MM 180 (viola, measure 25): these
are, respectively, one-tenth and one-half of MM 360.
The accumulation, as I have termed it, of a set of speeds by such rela-
tionships is important to Carter's later work; at this point in his develop-
ment, however, the connection between speeds that express metric-modu-
latory relationships and the idea of "simultaneous streams of different
things going on together" is still not fully realized. There are plenty of
these simultaneous streams, and occasionally they are treated as specific,
sharply defined speeds, but not consistently, or for any great length of
time. Some of the reasons for this have alreadybeen mentioned: the high
degree of flux and the consequently frequent dissolution of stratifiedmate-
rials. Another crucialbarrierto the consistent employment of simultaneity
at this stage is the lack of reinforcingpitch aspects, such as intervallicor pc-
set usage that is integral to the stratification. Schiff has enumerated
"themes" for the two outer movements of the First Quartet, but in prac-
tice these have no very specific identities; further, most of them have no
specific associationwith a single part but instead migrateabout the texture.
Quite a different state of affairsis found in the Second Quartet, eight years
later.
The significanceof measures312-50, however, should not be underesti-
mated. Carter has, after all, placed this passage as the climax of the first
movement-"the goal," as he has said, "of the techniques of metric mod-
ulation which have been used."22 In other words, the extended use of
simultaneous speeds in these measures is explainedby all that precedes it:
the juxtaposition of more or less briefly maintained speeds which leaves in
large part an impression of successive development. That is, speeds grow out
of one another, often in very rapid order, to be sure, but successivelyjust
the same. A similar process of explanation could be said to occur in the
Adagio movement (the third of Carter'sreal movements, which takes place
within the second literal one, as measures 57-154). Here the dark, rhyth-
mically complex, and passionate music of the viola and cello is contrasted
with the remote, high dyads of the two violins, moving serenely in long
values. Not until measure 85, however, do these two elements sound
together; before this point they are set forth in separatesections (measures
57-66, 66-76, 77-85). In both the Fantasiaand the Adagio movements,
nevertheless, the force of development is aimed at producing, in the end, a
condition of simultaneity, suggesting that the idea has alreadytaken on an
importance in Carter's aesthetic which will lead eventually to its employ-
ment in a truly integral role.
178 Perspectives of New Music
r I
L ,
n FE \
r 4~
b- --
^ --| I
,
AI% I { 1
4
-il
o-,
I ^*^ *^*^i L-,
ElliottCarter'sRhythmicPractice 179
'
I)- ?
I
) - ~
-~p1
wp -, S
Sf
_-?
-4
w?
I A )
L.," 0
-
A%
V *
O
frs~~ ;i, > 5
L o)"*t' llf)- c
_ '
|
180 of NewMusic
Perspectives

Moestoso J - 72
MM 96
.41 \

- - - - r
/t i z \

72 x 5 = MM 360 MM 48 farrache, rAorc

it
I*"~' r

^^
u^ T, Jl- "0

(.n.-72)h (1- 120)

).=A
if'2 (JclZo)

I.,
nO--72) - )' e )= * { 120)
---

--~ - rL-
I
- O.
ml- ---
-" T IL~

EXAMPLE 9: FIRSTQUARTET, I, MM. 9-21


ElliottCarter'sRhythmicPractice 181

VARIATIONSFOR ORCHESTRA (1954-55)

Carterreports: "In the course of exploring metric modulation, the idea of


dealing with accelerandos and ritardandos intrigued me. The first nota-
tional solution of an accelerando, which speeds up regularly from begin-
ning to end of a piece, occurred in the sixth variation of my Variationsfor
Orchestra... ."23 Complementing the sixth variation is the fourth, which
treats ritardando in a similar manner. The notational technique is related
only indirectly to that of metric modulation, as can be seen from Example
10 (measures289-99). The composer comments:

... a scheme of six measuresin %time speeds up during its course to


three times its initial pace [J = 80 to J = 240], at which point there
is a switch of notation, and a part previously playing quarternotes is
written in triplets of eighths, while in other parts dotted quarters
become eighths, dotted halves become quarters, and eighths become
sextuplets of sixteenths. Yet, while each of these notational systems
sounds as if it were continuing a regular acceleration, the beat has
returned to the speed of that of the first beat of the six-measure
scheme. The entire variation is projected onto this scheme, which
repeats itself over and over.24

Notice the entrance of the cello in measure 295, as if in canon to the clar-
inet at the tritone below. Carterthus achieves a gradual, almost completely
smooth acceleration, in contrast to the shifts of speed in metric modula-
tion, which occur as a series of distinct (even if somewhat de-emphasized)
steps. Later in the variation than the passagedisplayed in Example 10, the
"other parts" mentioned above by Carterenter, accumulatingas the varia-
tion progresses. The acceleration leads, analogously to the movement of
the stripe on a barber's pole, to a resumption of the original speed; but
from the beginning of the process, and increasingly as the variation pro-
ceeds, the overlap produced by parts that continue to accelerateeven as the
originalnotated tempo is re-establishedsucceeds in blurringthe boundaries
of the six-measurescheme. Furthermore, the thematic strand used consis-
tently throughout begins at some points at speeds slower or faster than
MM 80, thus adding several dimensions to the overlapping process. For
examples of this technique, see the third entranceof the thematic strand at
measure 298 (oboe) in Example 10 and the fifth and sixth entrances at
measures 307 (bassoon) and 308 (solo violin) respectively (not shown).
There are further elaborative features which deserve mention. First, as
each strand graduallypeters out, it decelerates against the current of con-
stant acceleration, sometimes establishing a more or less constant speed
before finally ceasing. Second, the thematic strand that begins the variation
182 Perspectives of New Music

NOTE for the CONDUCTOR:


The occelerondo extending over each group of six measures
should be a very regular speeding up fromJ = 80 to three
times that speed each time. Figures,such as that of the Clarinet
at measures 294-295, which extend over the double-bar and
change from quarters to triplets of eighths, should sound as if
they were continuingthe accelerando in quarters.

VARIATION 6
l289D Accel.molo l- - . - -
J80
J^ 96 .15 J. r39s 166

~L)~r
BkCL
4.
gbCL
.^
1 ~i J4
p lb
4-
1
bJ
c b ~ 1 J sy
\ ^

soloto

o
Sub.TEMPO I. -molto accel. -
|2951

.),J, B0 - J--96 J 115 r^ 139 J.166

08 1. -

B I
-CLf -[/w. _

.r
c|elloso,lorrsP hJlrrt j I P
tr ( f

EXAMPLE
10: Variatiosfir Orbhestra,MM. 289-99

appears in recognizable form throughout, but in the course of its multi-


plication it produces mutated versions of itself which are heard either as
continuations of the strand (or some abbreviationof it) or as isolated frag-
ments. In measures 315-38, for example, seven different string parts enter-
ing at separatepoints project closely relatedfragmentsin various schemes of
deceleration.
In all, the thematic strand along with its derivatives is heard in at least
two and as many as six of its phases simultaneously throughout the entire
variation (except, of course, during the first six measures). The precisely
controlled accelerationand retardationemployed here may be thought of as
special devices in Carter'swork. In this instance, however, the arrivalat a
scheme of simultaneous versions of identicalor similarmaterialthrough the
applicationof a single, thoroughgoing concept representsa new and power-
ful kind of compositional control that would permit in subsequent works
the definition of simultaneity as the essential fact from which sense and
structure devolve.
ElliottCarter'sRhythmicPractice 183

SECOND STRING QUARTET (1959)

In his published conversation with Allen Edwards, Flabtd Wordsand Stub-


bornSounds,Carter described his Second String Quartet as follows: "The
total notion of the piece is derived directly from the idea of simultaneously
interacting heterogeneous character-continuities."225 On the whole, it
seems significantthat this work, where Cartermade simultaneity the basis
of structure for the first time, also marks the first wedding of rhythmic
aspects with pitch aspects of distinction between parts. It is as if the eleva-
tion of simultaneity to its level of importance in the Second Quartet could
not have been accomplished without the mutually reinforcing power of
pitch and rhythmic considerations. Some of the details are attested to by
Carterhimself:

Here the four instruments are stratifiedaccording to their repertoire


of intervals, their repertoireof rhythms, and their repertoireof musi-
cal gestures. The first violin, for example, specializes in the intervals
of the minor third, the perfect fifth, major ninth, and major tenth.
Its fantasticand ornate characteris borne out by its rhythmic reper-
toire, which is extremely contrasted. The second violin, on the other
hand, shows very regularmotion and moves steadily at its own met-
ronome markings of 140, 70, and sometimes 280. The viola spe-
cializes in rhythmic relationshipswhich are usually in the ratio 2:3 or
3:5, and the cello does not move at a steady tempo, but rather has
accelerandosand ritardandosbuilt in.26

Like the first violin, each of the other instruments does indeed have its
own "repertoire of intervals," even though Carter has not mentioned
them here. This feature of structure is displayed in Example 11 (measures
45-50); accompanyingthe musicalillustrationis a tabulation of the relevant
intervals. Notice that a distinction has been made between the tritone, or
[6], in the viola and the compound tritone, or [18], in the cello. Majorand
minor seconds, on the other hand, are freely used as melodic connectors in
all parts.27 Thus it is clear that some elements are held in common; the
intent is not to divorce each part entirely from the other three. The reper-
toires enumerated by Carter and supplemented in our Example 11 tend to
produce what he would call "characteristicbehavior," a formulationwhich
obviously allows exceptions. The limits of behavior defined in this fashion,
in other words, are not at all absolute. One notices, for instance, that the
cello does not alwaysaccelerateor ritard;it merely does so more often than
the other instruments. And even among the intervalswhich are not major
and minor seconds, not every interval in a given part belongs to the reper-
toire established for that part. Sometimes, in fact, the entrance of one
184 of NewMusic
Perspectives

3 7 37 7 3 14 3 7 377 3

e
| ' g
tl "^3 ^ f =i B
J' '
? arco 5

f-~'atL,"UI- Ff
P10 13
fJr
-no f 13 f.ma. mp ,eo8p1s 1--0

5 18 5 18 88 5 8 5 5

3Y3 J17 73 3

6---, 10-3 6 - l ^^^r-u


l^rjP^^i^r It^*i~I~J' Y
won rpo
^__ ,. ''
o,.or ,~-~ ,L
------,k .....
..k~'
-7 i^T^"
,_,^jrS r3- r ^ -
I

8 5
fk 18 85
8 5
ri = important secondary part
wichtige Nebenstimme

intervalrepertoires

violin I: 3, 7,14, (16)


violin II: 4, 9, 11
viola: 6, 10, 13
cello: 5, 8,18

EXAMPLE 11: SECOND QUARTET, MM. 45-50


ElliottCarter'sRhythmicPractice 185

instrument will complete an interval, with a note already sounding in


another instrument, which belongs either to its own repertoireor to that of
the other instrument.28 The situation becomes further complicated as the
work progressesand each part begins borrowing and acquiringvarious traits
of the others.
As the separate "personalities" of the four parts begin to exert mutual
influences, it becomes clearthat pitch and rhythm remain intimately associ-
ated for the purposes of expressing the recombination and fusion of partic-
ular traits. The beginning of the fifth section (Cadenza for Cello, measures
243-60; see Example 12) is an apt illustration of this principle. Carter has
said that in this section "the cello, playing in its romanticallyfree way, is
confronted by the others' insistence on strict time."29 In reality, however,
the characterof this section is not nearly so simple as Carter's brief sum-
mary, written for the liner notes of a recording, might lead one to believe.
The "insistence on strict time" entails the projection of not one but three
rhythmic patterns in opposition to the cello's rubati. Two of these, in the
two violins, are strict speeds of MM 93.3 and 140 respectively. Recall that
at the outset of the Second Quartet it was the second violin that was
assigned the role of strict timekeeper. In this respect it differs from all the
others, especially the cello, whose constant accelerandiand ritardandiare
the very antithesis of strict time. Here the second violin maintains the role
it has held from the start-and, further, does so at the central speed of its
original scheme, MM 140. However, its pairingwith the first violin is new,
and before long the two instruments begin to borrow each other's inter-
vals, as if in expression of common purpose. From the annotations in
Example 12 it is evident that the first violin borrows more frequently than
the second-a condition which corresponds to its greatly altered rhythmic
characterin this passage.
Meanwhile, the viola, like the two violins, begins this section with a
statement of strict speed (MM 124.6). Soon, however, comes an unmistak-
able move away from consistent presentation of this speed. At first (meas-
ures 252-55), the multiple values preservesimple relationshipsto the origi-
nal speed; here t. = 124.6, and the two durations in question are, in
order, three and four times the length of the dotted eighth. But beginning
with the pickup to measure 256, no such simple relationshipprevails. From
here on the viola treads a fine line between freedom and stricture, never
surrenderingentirely to either. For the most part it may be said to keep to
itself, and the maintenance throughout of its original intervallicrepertoire
is a reflection of this position. However, set in the middle of the extreme
contrast created by the other instruments the viola also, perhaps inevitably,
assumes some aspects of the role of mediator. For this reason an occasional
interval from either the strict side or the free turns up. (Note the major
sixth, measure 255; perfect fourth, measure 258; major third, measure
259.)
186 Perspectives of New Music

243 Caden-u fi)r Cello


= J - 93.3- Penza sord.

lj,}~~4J4<J4^
t^~~~~t -' y^~ dRt- | .. .......................
ixEw;~......
~~~
piZZ
rLp r 7t pp

mf (~)

iZa -
(con -ord.) y

t= - -
... ......
. . ............
<.
-enza coord

(con sord.) M V -
18 " -- ,- 4'yfpr Y
- = = JD .
,,3n ~ jm.l ............_..
--...............
ISY?-

ere.s. poco a poco

molto
yiusto b9
- P
lSma t) un POrO me no
|
[^ stace. corto ed inciso, aop. arco

f marato - un poco meno f


stacc.
iinciso,corto ed accomp.

Oi o -
.......

f ,spr. f ,, a
fum ben infrz
sempr .. cantando
intenso

EXAMPLE12: SECOND QUARTET, MM. 243-60


ElliottCarter'sRhythmicPractice 187

e f molto espr.

0 U
^ ' ' ' P !
^--rT ^ I
^ 7 ^
3 F

7 - "f> 7 '- *p ' <;,

..???????~??????????????????????????.
r 1,o--
ar
..........;;
r .A 5
...........................................................................................................
' '^ v "^ ' '
fi; p^ F ^ ?^ __jf 6

yI
'-1
1U 912026

^ 1.. PZ7.3 arco .?

I , .3 ,
C # a. j
?J r-eoC DJ
e - } $ # 5
P -PP P

!i ?_h /.---.|' L
?, .."-rI "-.-'''''." ?c:erznd?o '' /?'f ' natu-ral

p,p ,p=m-
,^ P=- pp1 mj y
M p
PP
5 "
P sub.
^.
188 of New Music
Perspectives

DOUBLE CONCERTO(1961)

The next step in Carter's rhythmic development is also the most drastic,
although it has been well prepared by what has gone before. The Double
Concerto, completed in 1961, marksthe first time in Carter'smusic that the
condition of simultaneous speeds becomes normativefor the compositional
fabric.At the same time, each speed is pairedwith a specific pitch entity-in
this case an interval-such that each independently moving line is dis-
tinguished in two dimensions at once. The method is clear from the very
beginning of the Concerto. As is well known, a texture often simultaneous
speeds graduallyaccumulatesin the Introduction, beginning at measure 7
and completed in measure 36; entrances are arrangedso that shortly after
the complete texture has accumulated, the attackswhich determine the dif-
ferent speeds coalesce into two successive groups of six simultaneous
attackseach, at the downbeats of measures45 and 46, then disperse again.
Some of the features of this remarkableconstruction have been exposed
in Carter'sarticle, "The OrchestralComposer's Point of View," originally
published in 1970 and since reprinted.30Included in the articleis a chart of
the speeds, reproduced as Example 13. Carter's aim here was to arriveat a
unified, simple expression which would also yield a wide variety of rela-
tionships. The bracketingratio 2:1, represented by the metronomic mark-
ings 17.5 and 35, is filled in by eight other speeds to complete a scheme of
ratios 5 through 10 and their reciprocals.The remaining relationships, of a
more complex sort, are also bracketedby the 2:1 ratio, as their fanlike dis-
position in the chart reveals. These control the order in which the ten
speeds are introduced: the first two are 24.5 and 25, in the ratio 49:50; the
next two are 217s and 28, in the ratio 25:32; and so on, moving outwards
from the center until 2:1 is reached.
This much is common knowledge, but the actual details of the scheme
as it is worked out in the composition have not received the attention they
deserve. The procedure followed, as it turns out, is not at all as straightfor-
ward as one might expect it to be from Carter'slucid description.31At this
point a document included among the Cartermanuscript and sketch mate-
rials on deposit at the New YorkPublic Library(transcribedhere in part as
Example14) becomes relevant. This is Carter'sdurationalplan for measures
7-48 of the Double Concerto, showing the entrancesof the speeds as they
occur one by one.32 (For the sake of clarityI have added the speeds in MM
numbers at the left margin in each system.)
One important detail in the diagram is the comment at the bottom of
the first page: "Only those notes markedwith 2 are played." By "notes,"
of course, Cartermeans durations: the groups linked by ties. Not all writ-
ten attackpoints have accent marks. For example, in the line corresponding
to the speed 24.5, beginning in measure 7, Carter has indicated the first
Elliott Carter's Rhythmic Practice 189

RATIO BETWEEN SPEEDS METRONOMIC PIANO HARPSICHORD


* * * SPEEDS

2 1/5 35-

81 9 31/2 -

25 I/6 29'/6

32 8 28

50 '/7 25 -- l-

49 7 24/2

25 I/a 217/8--ti*

18 6 21--

50 1/9 :. 19% i

'/,o 5 7 '/ 17~


~:~~

EXAMPLE 13: DOUBLE CONCERTO,


CARTER'S TABLE OF SPEEDS

two attacks, which determine the 24.5. The next two are labeled 12.25, or
half of 24.5, because the attack point in measure 9, on the fourth part of
the septuplet in beat 1, has no accent mark. This sort of omission, however,
does not represent a real problem in terms of the projection of the overall
scheme, even though it occurs fairly often. The explanatory notes accom-
panying the diagrams state that the speeds chosen for this work are also
U)
X

190 of NewMusic
Perspectives

w
IE m m m ?" x
ee
si _ r v o
> N> f;
;S
> rn _ > _
> e os
percussion entranceof
perfect4th - 28

~~~~~~~~~~~~~~~I
~~~~entrance
entranceof
major7th
21 7/8

All sketchesarecopyright1980 by ElliottCarter DIAGRAM#1


and areused by his permission. only those notes markedwith > are played.

EXAMPLE 14: DOUBLE CONCERTO, CARTER'S RHYTHMIC PLAN,


192 of New Music
Perspectives

used in multiples: multiples of two for the five speeds assignedto Orchestra
I, with harpsichord,and multiples of three for the other five speeds, which
are assigned to OrchestraII, with piano. That is, for example, MM 28 in
OrchestraI may be presented as MM 56, while MM 35 in OrchestraII may
occur as MM 105. Here in the Introduction, where the speeds are not yet
thoroughly established, Carter has taken the additional license of present-
ing them from time to time at half their "true" values. This assures that
continuity is preserved: when an "expected" attack does not materialize,
as in measure 9 in our discussion above, or instead receives the negative
articulation of silence, the speed in question can be said to have dropped
momentarily into the background. In more general terms, this procedure
corresponds to Carter's intentions for the entire work: "In a way, all the
materialis being sounded in vestigial form almost all the time in the back-
ground. The idea that there is always a large world going on from which
items are picked out, brought into focus, and allowed to drop back, is one
of the fundamental conceptions of the piece."33
Another noteworthy aspect of the diagramis the information it conveys
about the instruments that actually articulatethe speeds as they enter. In
measures10-11, for instance, the "entrance of major second" is indicated.
One might assume that the intervalwould appearat the accent right at the
end of measure10, but as Example15 shows, the attackis made in the third
percussion part (Orchestra II), with the first pitch of the major second
entering in the violin a full two quarterslater. In fact, accordingto the dia-
gram in Example14, this speed (MM 12.5, half of 25) enters initially at the
end of measure 6 and is carriedby percussion 3 to the end of measure 10.
Not only that, however, but the percussion continuesto carrythe 12.5/25
speed; the interval itself does not enter explicitly into the scheme but
instead remains "associated" by occurring, when at all, only in proximity
to the attacksin the percussion.
In order to understand how this is supposed to work, the analyst needs
to know that beginning with the Second Quartet Carter conceives the
largerdesign of each of his works according to some dramaticplan which
occurs to him in the process of deciding how best to handle a particular
ensemble-the string quartet, for instance-or to write in some given genre,
however broadly defined, such as the concerto. In the case of the Double
Concerto the relevant model is Lucretius's De rerumnaturae (Concerning
the nature of things), "in which the cosmos is brought into existence by
collisions of falling atoms," a process which is representedin the music by
the accumulation and projection of the ten speeds through the climax of
measures 45-46 and beyond.34 After this a process of progressive differ-
entiation takesover as smallernumbers of speeds are presented in successive
and overlapping combinations. But during the Introduction the music is
still emerging from "chaos"-that is, lack of differentiation. Carter has
ElliottCarter'sRhythmicPractice 193

(105 x 5/42 = 12.5)

S 5
| 2 1- SC) _

Perc.3 11 | J _- J 7 I
_ _ _ _ _ _ _ _ - -
P - _--
Tom l 21 Bgo2rM
I , -
Perc4
1aiJ YJ J, . J, J JI
I J . JJ
J I
P 'roSD lrrlj ( c P

Pno.

9:

Vln
1.4
W ~ ~ j^* Tm __'j
.m b
t
.17 -- BfI
- i

_ ~- 14 'r I

v-c-.
-vc.
?_----
-
*-1
J IF
g
Y-,&
v
-
pp _ - - 1-

EXAMPLE 15: DOUBLE CONCERTO, ORCHESlRA II, MM. 10-14

mentioned that the "gradually changing percussion sounds... first 'give


birth' to a few musical pitches that in turn bring on the sound of the piano
and then the harpsichord,which in their turn become more and more artic-
ulated and differentiated.... ."35 Thus the speeds that have been so meticu-
lously worked out do not yet, in the Introduction, engage the intervals in
the precision of their workings-not reliablyor consistently, at any rate.
Interesting things do happen almost from the first, however, aside from
the articulation of Carter's rhythmic plan as shown in the diagram. For
instance, two entrancesof the major second, both connected by proximity
to the main scheme (also shown in the percussion part), themselves articu-
late the speed MM 12.5 exactly, but independently of the main scheme (see
Example 15). Even though this particularprogressiongoes no further, it is
clearlyan indication of the first tentative movements of the materialtoward
precise rhythmic definition. In Example 16, the speed MM 12.5 is further
reinforced-but not by the major second. Instead, two presentationsof the
major seventh, E-D# in horn 2 and bassoon, are used. Example 15 shows
194 of NewMusic
Perspectives

EXAMPLE16: DOUBLE CONCERTO, BASSOON


AND HORN 2 (ORCHESrRA II), MM. 13-15

where the speed corresponding to the major seventh, MM 21s, first


enters: with the sfin percussion 4, at the end of measure 13. Obviously, at
this very earlystage the speed is not at all established, and Carterhas appar-
ently chosen to add to the impresion of primal chaos by using the major
seventh to define the wrong speed, as if that interval did not yet "know"
what its proper behavior is.
Carter's collection of analytical material for the Double Concerto also
includes a series of diagramsthat he made to illustratethe principalfeatures
of the slow movement (measures313-468). In his article, "The Orchestral
Composer's Point of View," Carter also mentions these features: slow,
quiet music played by a central group made up of instruments from both
orchestras, which proceeds seemingly without relation to a succession of
accelerandiand ritardandiproduced by entrances of the instruments that
are positioned at the periphery of the ensemble in its performancedisposi-
tion.36 Example 17, reproduced from the published score, shows how this
disposition is supposed to look.37 One possible interpretationof this seat-
ing plan would be to regard the winds as a center, with the percussion,
strings, and keyboards stationed along the edges. In Carter's diagrams of
the slow movement, one of which is transcribedin Example 18, this parti-
tioning of the ensemble is in fact realized. The passagediagrammed, meas-
ures 331-42, consists of accelerandionly (plus, of course, the centralmusic)
and is to be read at all points in counterclockwise direction (ritardandiin
other diagramsare notated clockwise). The correspondenceto events in the
score is exact, as the accompanyingrhythmic reduction (Example19) makes
clear.
Elliott Carter's Rhythmic Practice 195

There should be as wide a space between the two orchestras


as the stage allows:

-- .---- ORCHESTRA I ----- -c 4 >r - - - ORCHESTRA II-


Perc. Perc.

[Winds:
e< T'bI
Tbn. as much Bn.
Bn.
Perc. \S Pe. Tbn.o , separation
Hn. I Tpt. Fl. as ob. CI. Hn. 1I
possible] .
Cb. Via. Vn. Vcl.
[Strings evenly spaced ]

I Cond.

0 no.

EXAMPLE 17: DOUBLE CONCERTO, SEATING PLAN

339

EXAMPLE 18: DOUBLE CONCERTO, MM. 33142,


CARTER'S DIAGRAM OF ACCELERANDI
196 of New Music
Perspectives

( sempre accel. )

X = 92 -- ------- -105 - 121 - - 139-

Pa.,Vc. Pa.

( semprle accel. )
d. == 70--- --------- ---- --- ------------------ ---
159--------------1S3 -----------------(210) p3 Pa.
a 3
r---iu

P.3

P.4 P.2

=70
92 ----------------- 06----------------- 122-----------------(140)
P.3 P.1 3

m !A^$
^17mm
- irr$. 12 _ 7
I
LL 11;
P4LL LLI L -
PLL
-.2
P.2 p

EXAMPLE 19: DOUBLE CONCERTO, MM. 331-42, RHTrHMIC REDUCTION

Meanwhile, what happens to the speed identities of the intervals?Accel-


eration and retardationdo not exclude the possibility that each successive
interval (or pairof intervals, in the case of verticalpresentation)could occur
at a different speed yet still fit the range of speeds established for it at the
outset of the work. Constant speeds, however, are ruled out by definition;
this and the difficultiesof calculationat a constantly changing speed may be
the reasons why interval identities are suspended in the peripheral parts
during these sections. The same difficulties apply to the "quiet music,"
but, perhaps because these parts are deliberatelydesigned not to accelerate
or decelerate, the assigned speeds are at least approximated. In Example 20
there are two demonstrations of this: the minor seventh at about MM 35
and the minor third at about MM 19.5 (that is, very close to 194/9).

_J J80. _ _ _J..106J-
- - _ _ _ _ - - - -J _2 _
r, - -1 r, -- -1I
Picc.
approx. 35 ;f. - ==. - t
Hn.1
approx. 19.5 M - -
Tpt.

Tbn. ~): -(" -'1 r


I$p,, -
p, mo,,oqt-

EXAMPLE 20: DOUBLE CONCERTO, PICCOLO


AND HORN 1 (ORCHEsrRA I), MM. 337-41
ElliottCarter'sRhythmicPractice 197

CONCLUSION

The projection of simultaneous speeds together with accelerationand retar-


dation in the same passage represents a kind of culmination in Carter's
rhythmic practiceup to this point. This accomplishment, of course, hardly
signaled a halt in Carter's development. His next work, the Piano Con-
certo, tackleswhat is in some ways an even more formidabletask: composi-
tion with all twelve trichords, each of which has a set of speeds.38 Carter's
shift in this work from intervals to trichords as the principal pitch-struc-
tural components may reflect, not a lack of satisfaction with the Double
Concerto, but rathera desire to attempt a different solution to the problem
of wedding pitch and rhythm on a formal level. The other fundamental
aspect of pitch structure in the Double Concerto-the assignment of the
two all-intervaltetrachords, one to each orchestra-is also part of this solu-
tion and deserves comment at this point. There would seem to be a certain
inconsistency in assigning, on the one hand, an all-interval entity to each
orchestra, distinguishing one part of the ensemble from the other on that
basis, and on the other hand in assigning five different intervals (not inter-
val classes)to each orchestra. In fact, Carter found it necessary, as he
explains in his unpublished notes on the Double Concerto, to add the tri-
tone to the piano's repertoire of intervals and the minor sixth to the
harpsichord'srepertoire, without, however, identifying them with particu-
lar speeds.39 In both cases the added intervalsupplies the one intervalclass
missing from the collection of five with assigned speeds. Each orchestra's
collection of intervals, then, represents the entire universe of interval
classes, yet the way in which this universe is constituted is in each case dis-
tinctly different; that the two all-interval tetrachords are different
tetrachords,after all, reinforcesthis distinction. However, the fact remains
that intervalclass is significantin some ways but not in others. This dichot-
omy between interval and interval class persists in Carter'slater works, but
with a different meaning, one which suggests more strongly than does the
dichotomy of the Double Concerto a globality of conception.40
American music owes a great deal to the fact that Carterwas seized, at a
crucial time in his career, with the same desire for rhythmic freedom that
had infected Ives, Cowell, and Nancarrow. At least as important, however,
is that Carter took a different path to that freedom, and in doing so man-
aged to escape the painful appellations of "visionary" and "experimen-
talist" which have come to represent the isolation, even belittlement,
accorded his predecessors. In barely more than a dozen years Carter
brought about two major changes in his music: first, a method of control
that allowed him to define rhythmic activity on multiple levels while show-
ing clearlyand audibly the interrelationof these levels; second, a concomi-
tant focusing of the pitch domain, with the result that definition of struc-
ture in terms of intervallicusage became fused with definition in terms of
198 of New Music
Perspectives

rhythmic usage, in particularthe projection of both simultaneous and suc-


cessive speeds. From the Double Concerto on one reallycannot say that the
rhythms conveythe pitches without adding that the reverseis at least as true.
In fact, in an interview conducted in 1968 Carter objected to his inter-
locutor's referenceto matters of rhythm, meter, and duration in his music
as "presentational," saying: "I consider [the rhythmic and metric aspect]
intrinsic to my music, just as intrinsic as pitch. '41 ClearlyCarteris justified
in making this claim. And while he did seem to arrive, around 1960, at the
"solution" to his problem of integrating rhythmic innovation with the
pitch structures of his music, the course of his career since then demon-
strates that this solution is hardly formulaic: that it has led Carter to an
effectively infinite compositional flexibility practicallyguarantees that his
music will continue to fascinatetheorists and composers for a long time to
come.
ElliottCarter'sRhythmicPractice 199

NoTEs

1. Elliott Carter, "The Rhythmic Basis of American Music" (1955), in


The Writingsof Elliott Carter (hereafter:Writings), ed. Else Stone and
Kurt Stone (Bloomington: Indiana University Press, 1977), 160-65.
2. Ibid., 163-65.
3. Carter, liner notes for Nonesuch recording H-71234 (1969), in Writ-
ings, 270.
4. Conlon Nancarrow, Rhythm StudyNo. Ifor PlayerPiano, New Music
Quarterly 25, no. 1 (October 1951): 1-22. According to Rita Mead
(Henry Cowell's"New Music" 1925-1936: The Society,the Music Edi-
tions, and the Recordings[Ann Arbor: UMI Research Press, 1981],
369-78), Carter was on the editorial board of New Music Quarterly
continuously during the years 1945-58 but was never chairman. Car-
ter himself, however, recallshaving succeeded FrankWigglesworth as
chairman, perhaps during the early1950s (letter to the author, 11Sep-
tember 1985). Mead has Vladimir Ussachevsky succeeding Wig-
glesworth in fall 1951 (p. 372).
5. "The Rhythmic Basis of American Music," 165.
6. Carter, "Shop Talk by an American Composer" (1960), in Writings,
209.
7. Carter, "The Case of Mr. Ives" (1939), in Writings,48-51.
8. See Henry Cowell, New Musical Resources(reprint ed., New York:
Something Else Press, 1969), 43-108 passim.
9. Carter, "Music and the Time Screen" (1976), in Writings, 346.
10. Allen Edwards, FlawedWordsand StubbornSounds:A Conversation
with
Elliott Carter(New York:Norton, 1971), 90.
11. Edwards, 101.
12. David Schiff, The Music of Elliott Carter (London: Eulenburg Books;
New York:Da Capo Press, 1983), 24 et seq.
13. Credit for coining the term goes to Richard Goldman, who was the
first to describe metric modulation in print in his article, "The Music
of Elliott Carter," TheMusicalQuarterly43 (1957): 151-70. The term
"metric modulation" is actuallya misnomer, since the meters,or writ-
ten time signatures, are only the agents of change, not the objective.
The realobjective of a "metric modulation" is a speedmodulation, and
200 of NewMusic
Perspectives

speed is a characteristicwhich is independent of notated meter (see


also note 14 below). However, because the term has stuck, and
because even Carterhimself now uses it, I follow the general practice
here.
14. Throughout this paper I often use the word speedto referto the rate at
which a series of (usually) identical values is moving. This can be
expressed as a metronomic, or MM, number. Speedcan also mean,
simply, tempo,but the latter term by itself is usually meant as a per-
formance indication-an actual, prescribed best rate-which is
obviously not the sense intended when one says, for instance, that six-
teenths at J = MM 120 move at a speedof MM 480.
15. " Music and the Time Screen," 349.
16. Benjamin Boretz, "Conversation with Elliott Carter" (1968), Perspec-
tivesof New Music 8, no. 2 (Spring-Summer1970): 19.
17. Schiff has revealed this order of composition (p. 132).
18. My analysis differs significantly from Schiffs (see p. 137). Both 4-14
and 4-17 are included in his tabulation of four-note subsets of a con-
trolling six-note set (6-Z43 [012568]), but 5-11 apparently does not
figure in his scheme, since it is not a subset of 6-Z43.
19. "Music and the Time Screen," 356.
20. Schiff has also noted this intermovemental link (p. 141).
21. As Carter himself has noted, the literal divisions into movements of
the First Quartet do not correspond to the real movemental scheme:
the former is a tripartitearrangement,while the latter is quadripartite.
"The reason for this unusual division of movements," he has written,
"is that the tempo and characterchange, which occurs between what
are usually called movements, is the goal, the climax of the techniques
of metrical modulation which have been used. It would destroy the
effect to break off the logical plan of the movement just at its high
point. Thus pauses can come only between sections using the same
basic material. This is most obvious in the case of the pause before the
movement marked Variations [that is, the third]. In reality, at that
point the Variations have already been going on for some time."
("The Time Dimension in Music" [1965], in Writings,246.)
22. See note 21 above.
23. "Music and the Time Screen," 356.
24. Ibid., 356-57.
ElliottCarter'sRhythmicPractice 201

25. Edwards, 101.


26. "The Time Dimension in Music," 247.
27. Numbers in square brackets refer to sizes of intervals in semitones,
without respect to order or transpositionallevel. That is, for example,
the pitch order G3-C4 is taken to be equivalent to C4-G3 (or G03-
C#4 or Ct4-G33, and so on), but not to G3-C3 or G3-C5. The list
given in Example 11 agrees with Schiffs (see p. 32 of his book) with
respect to all but one interval. Orin Moe's article "The Music of
Elliott Carter," CollegeMusic Symposium22, no. 1 (1982): 7-31, has a
similar list, but it is incomplete. William Brandt, in his article "The
Music of Elliott Carter," reprinted in Breakingthe SoundBarrier:A
CriticalAnthologyof the New Music, ed. Gregory Battcock (New York:
Dutton, 1981), 221-34, has described a different intervallic distribu-
tion among the four parts. His findings, however, are in some
instances imprecise and in others simply wrong.
28. For this reason, it remains an open question whether the pitch struc-
ture of the Second Quartet can be adequately described with pitch-
class sets. In 1959, Carter's tabulation of pitch sets in the Harmony
Book was still severalyearsin the future, and it may well be that at this
point Carter had not consciously formalized relationships involving
groups largerthan dyads. There is some evidence opposing this spec-
ulation. Carter has stated that he used one of the all-interval
tetrachordsas a "key" chord in the First Quartet (1951): "This chord
is not used at every moment in the work but occurs frequently
enough, especially in important places, to function, I hope, as a for-
mative factor." ("Shop Talkby an American Composer," 205.) There
is also Carter's postcard to Michael Steinberg, dating from 1960,
which displays both all-interval tetrachords and comments that the
Second Quartet "is all made up of interval combinations derived from
these two 4 note chords-in all movements"; and, further, that the
First Quartet uses only the second of these-that is, [0146].
(Reproduced in Elliott Carter: Sketchesand Scoresin Manuscript [New
York:The New YorkPublic Libraryand Readex Books, 1973], 29.) In
their efforts to confirm these statements through independent investi-
gation, however, analysts have not yet gotten very far. Schiff, for
example, outlines the occurrences of [0146] only in the opening cello
solo of the First Quartet and does no more than mention the two
tetrachords of the Second. Theorists' general perplexity concerning
pitch structure in these works is well voiced by Carlton Gamer in his
review of FlauedWordsand StubbornSounds(Perspectives of NewMusic 11,
no. 2 [Spring-Summer1973]: 146-55).
202 of New Music
Perspectives

29. "String Quartets No. 1 (1951)and No. 2 (1959)," in Writings, 278.


30. "The Orchestral Composer's Point of View" (1970), in Writings,
282-300.
31. Rudolph Kompanek, in an unpublished master's thesis completed at
Eastman in 1972, has treated the Double Concerto in considerable
detail. He too has examined the Introduction and has found dis-
crepanciesbetween the music and Carter'swritten account of it. How-
ever, Kompanek did not have access to certain information which has
surfacedamong the sketch and manuscript materialson deposit at the
New YorkPublic Library,and his findings, for this reason incomplete
in certain respects, led him to conclude-erroneously, I believe-that
Carter'sstated plan does not work as the composer intended it to.
32. "Diagram 1," in a folder headed "Double Concerto: Diagrams and
Analyses Made in Early60's." The diagram, which I have abbreviated
in this transcription, actually continues through measure 52.
Although the hand is without question Carter's, the plan cannot
properly be called a sketch, for it was apparentlymade sometime after
the piece was finished.
The transcription in Example 14 corrects the line location of the
MM 24.5/12.25 speed in measures 7-11, omits a few markingsof only
incidental importance, and also omits (in measures 12-16) a large
amount of crossed-out materialthat was apparentlyintended to show,
before Carterthought better of it, how the other speeds in the scheme
of ten tacitly progress before they actually enter.
33. Boretz, 7-8.
34. "The OrchestralComposer's Point of View," 292.
35. Carter, liner notes for Nonesuch recording H-71314 (1975), reprinted
as "Double Concerto..." in Writings, 326.
36. "The OrchestralComposer's Point of View," 295.
37. Carter, Double Concerto,score (New York: Associated Music Pub-
lishers, Inc., 1964), iii.
38. See my article, "Spatial Sets in Recent Music of Elliott Carter," Music
Analysis 2 (1983): 5-34, in which I discuss some of the aspects and
implications of this technique as it is used in the Piano Concerto, par-
ticularlythe first movement.
39. Carter, "Double Concerto," one-page note, in "Diagrams and
Sketches."
ElliottCarter'sRhythmicPractice 203

40. The nature of globality in Carter'srecent music is addressed at length


in "Spatial Sets" (see note 38).
41. Boretz, 18. Boretz apparently does not view such a state of affairsas
possible in music, to judge from his assertionsthat "a musicalduration
is necessarilythe duration of something"and that "while durations are
independently specifiable, the question [of] which durations it is rele-
vant to specify hangs on a prior identification of the things of which
they are durations" ("In Quest of the Rhythmic Genius," Perspectives
of NewMusic 9, no. 2/10, no. 1 [1971]:152). No doubt music does exist
for which Boretz's statements hold true-perhaps, for instance, the
music of Stravinsky, Boretz's point of departure in this article. But
Carter's work since 1960 is more in line with what John Rahn envi-
sioned in his response to "In Quest of the Rhythmic Genius": music
in which not only do pitches "provide a framework for the articula-
tion of time" but also times are chosen which "provide a framework
for the articulationof pitches"; and, further, "in which bothprocesses
apply, each independently coherent and creating together a new, more
complex structure of counterpointed independent structures"
("Rhythm, and Talk about It," Perspectives of New Music 15, no. 2
[Spring-Summer1977]: 238).

Você também pode gostar